Systematic design of chemical oscillators. 5. Bistability and oscillations

Jan 1, 1982 - ... load: https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js .... For a more comprehensive list of citations to this art...
3 downloads 0 Views 653KB Size
J. Am. Chem. SOC.1982, 104, 504-509

504

The value of [B], the total base concentration, is found to be 9.0 X lo”, or pH = 9.1. This corresponds to a catalytic rate constant of 31 X s-l a t 25 OC, or 4 times larger than the 7.1 X s-l calculated at pH 8.6. Thus only a modest improvement in catalytic efficiency can be hoped for by optimizing conditions. Fe(C0)5 is a poor catalyst because the two steps involved in the catalytic cycle have exactly opposite pH requirements. It may be noted that Fe(CO)5 is not

a catalyst for the WGSR in acid media,31 as expected from our results. Acknowledgment. This work was supported by the National Science Foundation, Grant No. CH-77-24717. (31) Ford, P. C.; Ungermann, C.; Landis, V.; Moya, S.;Rinker, R. G.; Laine, R. M., Adu. Chem. Ser. 1979, No. 173, 81.

Bistability and Oscillations in the Autocatalytic Chlorite-Iodide Reaction in a Stirred-Flow Reactor’ Christopher E. Dateo, MikMs OrbPn,2 Patrick De K e ~ p e rand , ~ Irving R. Epstein* Contribution from the Department of Chemistry, Brandeis University, Waltham, Massachusetts 02254. Received July 14, 1981

Abstract: The reaction between chlorite and iodide in acidic solution (0.08 < pH < 3.75) has been studied in a stirred-tank reactor at 25 OC over a range of flow rates, pH, and input chlorite and iodide concentrations. The system is found to exhibit bistability and, at high [C102-]and [I-], sustained oscillations as well. The key reaction which gives rise to these phenomena is C10, + 41- 4H’ = Cl- + 21z + 2Hz0, which is autocatalytic in iodine and is inhibited by iodide. The existence of bistability and oscillations in this system is consistent with the “cross-shaped phase diagram” model of Boissonade and De Kepper.

+

The recent report of the first systematically designed homogeneous chemical oscillator4 emphasizes the enormous amount that may be learned about the dynamics of complex chemical systems from the study of autocatalytic reactions in a continuous-flow stirred-tank reactor (CSTR). In the above mentioned study, two autocatalytic subsystems, arsenite-iodate and chlorite-iodide, were linked together in a CSTR to give rise to oscillation. The existence of sustained oscillation in that system had been suggested by a model analysis which implied that addition of an appropriate feedback reaction to a system which exhibits bistability in a CSTR should lead to a “cross-shaped phase diagramn5in which oscillations appear for a range of input reactant concentrations. The arsenite-iodate reaction has been thoroughly examined in a CSTR6v7and has been found to exhibit bistability under a wide range of conditions. A simple model involving several overall component processes6has been shown to give excellent agreement with the experimental observations. Continuing our study of autocatalytic reactions in the CSTR, we report here on the remarkable richness of dynamic behavior exhibited in a flow reactor by the autocatalytic reaction between chlorite and iodide. We have observed not only the expected bistability but also sustained oscillation; the chlorite-iodide system itself is characterized by a “cross-shaped phase diagram”. In addition, as we report elsewhere, this reaction forms the basis of an entire new family of oscillating reactions when chlorite- and iodine containing species (IO3-, I-, or 12) are combined with a wide variety of reducing’s4 or oxidizing* agents. (1) Part 5 in the series Systematic Design of Chemical Oscillators. Part 4: Orbin, M.; De Kepper, P.; Epstein, I. R.; Kustin, K. Nature (London) 1981, 292,816-818. (2) Institute of Inorganic and Analytical Chemistry, L.Eijtvb University, H-1443, Budapest, Hungary. (3) Centre de Recherche Paul Pascal, Domaine Universitaire, 33405 Talence, France. (4) De Kepper, P.; Epstein, I. R.; Kustin, K. J. Am. Chem. Soc. 1981, 103, 2133-2134. (Part 2). (5) Boissonade, J.; De Kepper, P. J . Phys. Chem. 1980, 84, 501-506. (6) De Kepper, P. Epstein, I. R.; Kustin, K. J. Am. Chem. Soc. 1981, 103, 6121-6127. (Part 3). (7) Papsin, G.; Hanna, A.; Showalter, K. J . Phys. Chem. 1981, 85, 2515-2582.

Experimental Section The apparatus employed consists of a thermally regulated stirred tank glass (Pyrex) flow reactor, which has been described in detail elsewhere.6~~ Potentiometric measurements could be made by using either an Orion iodide-specific electrode or a platinum redox electrode with a Hg/ Hg2SO4/K2SO4electrode as reference. Simultaneous measurement of the absorbance of the solution in the CSTR could also be made. Absorbances were generally monitored at 460 nm, the wavelength at which the I2 extinction coefficient is a maximum. The path length through the cell was 3.5 cm. The constraints which are at the disposal of the experimenter are the temperature, which was maintained at 25.0 f 0.1 ‘C in this series of , the conexperiments, the residence time T (or flow rate /ro= l / ~ )and centrations [Ailo that the input species Ai would reach in the reactor if no reaction were to occur. The input chemical species of importance in these experiments are [CIO,-],, [I-],, and [H+l0 Chorite and iodide solutions were prepared by using doubly distilled water from the highest purity commercially available grades of NaC102 and KI. Reagent solutions were stored in three separate reservoirs. One contained the chorite, stabilized by 0.001 M sodium hydroxide. The iodide solution was prepared in either sodium acetate or sodium sulfate, and the third reservoir contained the corresponding acid, acetic or sulfuric, depending on the pH desired. Thus when the solutions were mixed in the reactor, an acetic acid-acetate or sulfate-bisulfate buffer of the appropriate pH was generated in situ. The pH was measured directly in the output flow.

The Chlorite-Iodide Reaction in Batch When solutions of chlorite and iodide are mixed in the absence of flow (bath conditions), an initial slow increase in iodine absorbance soon gives way to a rapid exponential rise. If chlorite is in excess, this rise terminates in an abrupt drop-off, with the disappearance of all trace of iodine color. The iodide concentration is found to decrease slowly at first and then to fall suddenly by more than 5 orders of magnitude at about the time of the peak in I2 absorbance. This behavior is illustrated for a slight stoichiometric excess of chlorite in Figure 1. With still greater excesses, the absorbance drops to and remains at an essentially zero value. (8) Dateo, C.; Orban, M.; De Kepper, P.; Epstein, I. R., to be submitted for publication. (9) Pacault, A.; De Kepper, P.; Hanusse, P.; Rossi, A. C. R. Heb. Seances Acad. Sci. 1975, 281C, 21 5-220.

0002-7863/82/ 1504-0504$01.25/0 0 1982 American Chemical Society

J. Am. Chem. Soc., Vol. 104, No. 2, 1982 505

The Autocatalytic Chlorite-Zodide Reaction

l

.

o

E

0

(0

t c 0

u 0 c 0

$ (R

n

-

6-

U

8-

1-

I

Figure 1. Absorbance at 460 nm and iodide concentration for a batch reaction at 25 O C and pH 3.3 with initial concentrations [I-]i = 4 X lo4 M, [ClOTIi = 2.5 X lo4 M.

First studied qualitatively by Braylo some 1 5 years ago, this spectacular clock reaction has been thoroughly investigated in batch by Kern and Kim" and by de Meeus and Sigalla." The initial phase of the reaction is characterized by the stoichiometry of process A. In the presence of excess chlorite, the C 1 0+ ~ 41-

+ 4H+ = C1- + 212 + 2 H 2 0

b'

4 31

4t

I

L Q

(A)

iodine is oxidized further to iodate in process B. If iodide and 5C102-

+ 212 + 2H20 = 5CI- + 4103- + 4H+

(B)

iodate are present simultaneously, then process C is also of significance. The net reaction in the presence of excess chlorite is 103(R) = I/z[(A)

+ 51- + 6H'

= 312 + 3H20

2

(C)

+ @)I. 3~10,-

+ 21- = 2 1 0 ~ -+ 3C1-

(R)

Kinetic studies 11,12 have established that reaction A is autocatalytic in Iz and is inhibited by I-, with a rate law of the formI3

Reaction B is known to be quite rapid, but a rate law for this process has not yet been established. Process C is the venerable Dushman reaction,14 the kinetics of which have been studied by numerous workers with a variety of somewhat conflicting results,15 though a rate law of the form d[IO3-1 --- kcl[H+]2[I03-][I-] + kc2[H+]'[I03-] [I-]' dt

(2)

seems to have a great deal of evidence in its favor. The Chlorite-Iodide Reaction in the CSTR Bistabffty. Under batch conditions, the final equilibrium state attained by a mixture with the initial composition shown in Figure (10) Bray, W. C. Z . Phys. Chem. 1906, 54, 131-149. (11) Kern, D. M.; Kim, C.-H. J . Am. Chem. SOC.1965,87, 5309-5313. (12) De Mceus, J.; Sigalla, J. J . Chim. Phys. P h y s d h i m . Biol. 1966, 63, 453-459. (13) The absence of the KA term in the rate law reported in ref 11 is compatible with the values of t6e rate constants given in ref 1 1 and 12 and with the much higher iodide concentrations employed in the experiments of ref 12. (14) Dushman, S. J. J . Phys. Chem. 1904, 8, 453-482. (15) Liebhafsky, H.A,; Roe, G. M. Int. J . Chem. Kinet. 1979, 11, 693-703.

4

10-4

10-3

2x

[I-]0 ( M I Figure 2. Steadv state hvsteresis ohenomena in the absorbance at 460 nm and the iodidk concentration in'the CSTR as a function of [I-I0,with [ClO